1

No.75

Design and Evaluation

of the Joint Venture Formation

in EPC Projects

Nobuaki Ishii, Kanagawa University, Japan

Yuichi Takano, University of Tsukuba, Japan

Masaaki Muraki, Tokyo Institute of Technology, Japan

December 8, 2018

APIEMS2018

Proceedings of the Asia Pacific Industrial Engineering & Management Systems Conference 2018

2

① Background & Objectives

② Related work

③ Design Method for Competitive Joint

Venture Formation

④ Mathematical Model in Design Method

⑤ Evaluation

⑥ Conclusions

Overview of the Presentation

3

EPC （Engineering-Procurement-Construction) Project

• Contractors design and build unique

products based on the client requirements.

• Contractor has a solo responsibility for the

project as a lump-sum contract.

Background About EPC Project

Typical Examples of EPC Project: Construction, Civil engineering,

Plant Engineering, Information System Development, etc.

4

Background About EPC Project

EPC contractors have been suffered

unstable business results, so far.

Contractors’ unstable business results

would also have

harmful effects on client’s business.

・Additional cost ・Delay of project delivery

(800,000)

(600,000)

(400,000)

(200,000)

0

200,000

400,000

600,000

800,000

1,000,000

1972 1977 1982 1987 1992 1997 2002 2007 2012 2017

年度

受注 売上 税引き前利益(百万円×10）

5

Background

Year

Example <Financial results of a world class contractor>

Accepted

orderSales

Profit (×10-1)

[MM¥]

About EPC Project

In EPC, contractor’s business performance is highly affected by the

market ＆competitors‘ conditions.

Stable profit is critical for any contractor to improve sustainability.

Unstable Sales & Profits

6

Background

Why so unstable?

Risks of cost & time affect profitability,

especially, in complex & large-scale EPC

projects.

“Reduce risks of project ” is critical for

the contractor to gain stable profit.

Risks & Profits

One solution:

The joint venture contract It has become more common among EPC contractors.

More than one contractor works on a project in the joint venture.

7

Background

Why Joint venture can reduce risks:

Each partner in joint venture needs not carry

out all the works in the project by sharing

works within partners.

Each partner only need to carry out the

works having specialty.

Risks & Profits

Risks of cost & time could be reduced by the

complementary effects within joint venture

partners.

8

Background

However,

Partners in the joint venture encounter many

challenges to gain advantages by forming a

joint venture.

<Examples>

Increased managerial costs,

Hard negotiations of profit & risk sharing

among partners, and so on,

Challenges of Joint Venture

A method to design a competitive joint venture

formation is necessary.

9

Research Objectives

Based on the background;

Develop a method to design a

competitive joint venture formation;

Reduce project cost and deficit risks,

by maximizing the complementary effects

within joint venture partners.

Related work

10

Related work

11

Most studies on joint ventures have focused on

complementarity in repetitive businesses,

Literature focusing on project-based business

has been limited so far.

Especially, issues for organizing competitive joint venture

formation in EPC projects have not been studied well by

either industry and academia, although expectations for

joint venture contracts have increased among EPC

contractors.

This paper focuses on the design of joint venture

formation in EPC projects

as a new area of joint venture studies.

12

Overview of design method

Overview of design method

13

The design method consists of four steps.

Estimated costs of WP by each partner

WP set conditions in each partner

Client requirement

s Create 2D-

WBS

Estimate cost

of WP (average

& variance) in

2D-WBS

Search

optimal

formation

Mathematical model for

optimization

Constraint of maximum projectcost

Cost data of each partner

2D-WBSsCreate 2D-WBS of

each partner

2D-WBS

Cost estimation

models

Joint venture formation

1st step

2nd step

3rd step

4th step

Overview of design method

14

1st Step:

Creates a 2D-WBS (Work Breakdown structure)

based on the project scope determined by the

client requirements,

The 2D-WBS:

Consists of a process WBS and functional

WBS,

Shows the association between the WPs of the

project and the potential partners carrying out

the WPs.

Indicates the work process in the

EPC project, such as design,

procurement, and construction

processes.

Shows the work breakdown to

deliver the results of the project.

C5

C4

C3

C2

D2

D1

C1

B1

B2

A

D1 D2 P1 P2 S1 S2 S3

Design Constriction

Process WBS

Procurement

Functional WBS

Each cell shows Work Package (WP) to

be done by partner

Work

Process

WPC3,P1

WPC2,S1

Two-Dimensional Work Breakdown Structure

(2D-WBS).

15

Final deliverable is A,

and the works to deliver A is broken down as C2 via C5,

and D1 and D2.

WP of work C3 in

process P1

Each joint venture partner has the same 2D-WBS,

although costs and those variances to carry out the WPs

would be different for each partner.

Tasks

Resources

Cost

Schedule

PartnerWPC3,S2

The 2D-WBS is used as a platform of joint

venture design in Steps 2 to 4.

Overview of design method

16

2nd Step:

Cost and MH data of each WP in the 2D-WBS is

estimated by each partner.

Cost estimation model & cost data would be

different for each partner,

Estimated costs & cost estimation accuracy can

be different for each partner. AACE cost model

Uses the data of the cost estimation accuracy to

evaluate the project deficit risk.

Cost estimation accuracy is defined as the percentage of

deviation from the actual cost.

Overview of design method

17

3rd Step:

2D-WBS of each partner is created by putting the

cost and MH data of WPs into the 2D-WBS.

Total N 2D-WBSs are created if the number of

partners is N.

4th Step:

Joint venture formation is searched based on the

project data in the 2D-WBSs by using the

mathematical model. The model finds the combinations of WPs and partners

who carry out the WPs so that the sum of cost variances

of WPs is minimized under the constraint of the maximum

project cost like a portfolio optimization problem.

18

Mathematical Model in Detail

Minimize

𝑛=1

𝑁

𝑤=1

𝑊

𝑝=1

𝑃

𝑎𝑤,𝑝𝑛 𝑑𝑐𝑣𝑤,𝑝

𝑛 (1)

Subject to𝑎𝑤,𝑝𝑛 ={0, 1} (2)

𝑛=1

𝑁

𝑎𝑤,𝑝𝑛 = 1 (𝑤 = 1~𝑊, 𝑝 = 1~𝑃) (3)

𝑛=1

𝑁

𝑤=1

𝑊

𝑝=1

𝑃

𝑎𝑤,𝑝𝑛 𝑑𝑐𝑤,𝑝

𝑛 +

𝑛=1

𝑁

(

𝑤=1

𝑊

𝑝=1

𝑃

𝑎𝑤,𝑝𝑛 𝑖𝑑𝑐𝑤,𝑝

𝑛 )𝑠𝑐 < 𝑡𝑐𝑜 (4)

𝑤=1

𝑊

𝑝=1

𝑃

𝑎𝑤,𝑝𝑛 𝑚ℎ𝑤,𝑝

𝑛 ≤ 𝐻𝑀𝑛 (𝑛 = 1~𝑁) (5)

If WPw,p and WPw’,p’ must be carried out by the single

partner n for all wand p.

𝑎𝑤,𝑝𝑛 = 𝑎𝑤′,𝑝′

𝑛 (𝑛 = 1~𝑁)

(6)

Minimizes the sum of

the variance of WPs’

costs.

Decision variable of the

formation of JV.

Constraints of project costs.

Constraints of MH of each

partner.

A set of WPs each partner

intends to carry out.

19

Numerical Examples

20

<Objectives>

Evaluate the effectiveness of the design

method of joint venture formation by

examining the expected profits of EPC

projects,

with the result of improved cost estimation

accuracy and/or the reduced project cost .

<Evaluation method>

Simulation model of the competitive bidding

is used for evaluation.

Numerical Examples

21

Simulation Model for Evaluating

Joint Venture Formation

22

Simulation Model --- Two modules

① Competitivebiddingmodule

Estimate costs

and bid prices

Expected orders & profits

Estimated costs & markups

(Bidding price)

Bidding price=Estimated cost +Markup.

② Bid evaluation

module

Evaluate bidding &

expected profits

Search the bidding price maximizes the expected

profits.

Consists of two modules.

Repeated

23

Design of Simulation Experiments

24

Simulation scenarios

Scenario OrganizationCost

estimationaccuracy

Cost reduction

S0 One’s own companyCompetitors +/-5% none

S1Joint venture

+/-5%0~5% of S0 cost

Competitors none

S2Joint venture +/-3% 0~5% of S0 cost

Competitors +/-5% none

S0 (Base scenario): No joint ventures.

S1: Reduced project cost by JV.

S2: Improved cost estimation accuracy & the reduced

project cost by JV.

Cost estimation accuracy = the percentage representation of the coefficient of variance, such as 5% of the actual cost.

It is obtained by dividing the standard deviation of the estimated cost by the actual cost. Namely, a lower deviation means higher

estimation accuracy.

25

Simulation data

Number of simulation runs SR 5

Number of simulation years N 5

Number of competitors

S0 5

S1 and S2 4

Number of orders in each simulation year 10

Probability function of project costNormal

distribution

Markup in competitor j: 𝑟𝑚𝑘𝑖𝑗 Random variable

between 9~11%

Actual project cost 100 [MM$/project]

Target volume of accepted orders in

each simulation year

Joint venture 500 [MM$/year]

one’s own company

250 [MM$/year]

(MM: Million)

26

Results of Simulation Experiments

0

5

10

15

20

25

30

35

40

0 1 2 3 4 5

S0 S1 S2

[MM$/year]

Project cost reduction in joint venture [%]27

The expected profits in the joint venture increasesaccording to the reduction of the project cost

The improved cost estimation accuracy by the effectof joint venture formation increases the expectedprofit.

Cost estimation accuracy: +/-3%

Cost estimation accuracy: +/-5%

Results of Simulation Experiments

Expected profits of accepted orders

Improved profits by the reduced No.of competitors

Improved

profits by

accurate cost

estimation

102

103

104

105

106

107

108

0 1 2 3 4 5

S0 S1 S2

[MM$/order]

Project cost reduction in joint venture [%]28

The average bid prices in the joint venture slightly

decreased according to the project cost reduction.

The joint venture can improve contractors’ profits as

well as reduce the investment costs of clients.

Results of Simulation Experiments

Average bid price of accepted orders

Reduction of client's

investment cost

29

Conclusions

30

Conclusions

①Develop a design method & a mathematical

model to identify a joint venture formation.

Minimizes the cost estimation errors associated with the

deficit risk of projects under the constraints on the

maximum expected project costs and MH.

Consider the complementary effects of the joint venture

and the managerial cost among contractors within the

formation.

②Propose the 2D-WBS as a platform to indicate

project data of each partner organizing the

joint venture formation.

31

Conclusions

③Findings via simulation experiments.

Joint venture can reduce deficit risks and

improve the expected profits

Joint venture can also reduce client's

investment cost

32

Conclusions

Further research

① The design method should consider more facets of

the joint venture,

Such as, knowledge creation, the presence of competition between joint

venture partners, and so on, to evaluate the diverse aspects of joint

ventures.

② The mathematical model should be expanded so

that it can consider the bid conditions and bid

strategy of each competitor,

③ In the simulation model of the competitive bidding,

the bid performance of each competitors should be

evaluated not as a set of competitors.

33

Thank you very much

for your kind attention.

This work was supported by JSPS KAKENHI Grant Number

16K01252.